도구

Lets agents drive the open-source CAD app FreeCAD: create and edit part models in natural language, run Python scripts, capture views and run FEM structural analysis for mechanical and product design. Needs FreeCAD with the addon and its RPC server, plus uv; engineers should verify results.

Lets agents work in KiCad directly: draw schematics, place components and route boards (with Freerouting autorouting), create footprints and symbols, and search the JLCPCB parts library. Needs KiCad 9+, Node.js 18+ and Python; clone and build the repo, then register it.

Search and download arXiv papers and read the authors' LaTeX source section by section, export BibTeX and follow topic updates, with papers cached locally. No API key needed.

Pluggable natural-language linter that checks Markdown and plain text for wording, punctuation and terminology consistency like code, with auto-fix, and can also run as an MCP server. Rules are installed separately, mostly for Japanese and English; requires Node.js 20+.

MCP server built into the textlint proofreader: agents check Markdown, plain text and other documents against your rules for wording, punctuation and formatting issues and get auto-fixed text. Needs Node.js plus a .textlintrc and rule packages in your project; textlint has no default rules.

Microsoft's official skill collection: 170+ skills focused on Azure SDKs (Python, .NET, TypeScript, Java, Rust) and Microsoft Foundry development, plus plugins, custom agents and MCP configs. Microsoft recommends picking skills per project rather than installing them all.

PDFMathTranslate 2.x (command pdf2zh_next), built on BabelDOC: translates papers and other PDFs while keeping formulas, charts, TOC and annotations, outputting translated and bilingual side-by-side versions. Uses a free translation channel by default, or your own OpenAI, DeepSeek or other API key.

Finance skills for US, Hong Kong and China A-share markets: news from 10+ sources, quotes and fundamentals, sentiment analysis, time-series forecasts, signal tracking, impact-chain diagrams and reports. Needs Python; some skills download PyTorch models. Not investment advice.

Unified search across 20+ academic sources including arXiv, PubMed, bioRxiv, Semantic Scholar and OpenAlex, returning deduplicated, normalized results and downloading full text via open-access channels first. Most sources need no API key.

Replaces arXiv preprint entries in your .bib file with the official published versions from DBLP and ACL Anthology, keeping citation keys, to tidy references before camera-ready. Overwrites the file by default, so preview with --dry-run. Mainly covers CS venues; needs Python 3.10+.

A JADX decompiler plugin plus companion MCP server that lets agents read an APK's classes, methods, Smali and resources to assist Android reverse engineering and vulnerability hunting. Requires JADX-GUI and uv.

Open-source, agent-friendly BI: define metrics and dashboards in YAML and SQL, and rill init generates project instructions and MCP config for Claude Code so agents can build business dashboards and answer questions in natural language. Local use is free; publishing to Rill Cloud needs an account.

Python library with a built-in MCP server for US SEC EDGAR data: search 10-K, 10-Q, 8-K and other filings and read XBRL-standardized financial statements, Form 4 insider trades and 13F holdings. Free with no API key; just set an EDGAR_IDENTITY.

A native x64dbg plugin that works as an MCP server, exposing dozens of debugging tools: agents can set breakpoints, step through code, read and write memory, analyze PE structure and locate the OEP of packed binaries. Windows only; install the plugin in x64dbg, then connect with a token.

Translates a finished blog post into one or more languages while keeping formatting, images and charts, localizes keywords, titles, meta descriptions, dates and currencies per market, and flags sentences that read like machine translation. No account needed; part of the claude-blog suite.

Bright Data's official MCP: web search (Google, Bing, Yandex), any page to Markdown, structured data extraction from e-commerce, social and other platforms, and remote browser control. Needs an API token; includes 5,000 free requests a month.

LLM translator for novels, subtitles, ebooks and game text: one-click glossaries keep names consistent and translations get auto-reviewed. Besides the desktop app it has a CLI (translate/analyze, JSONL status). Download from Releases and bring an LLM API; commercial use needs the author's OK.

Scans codebases and dependencies for open-source licenses, copyright notices and package manifests, detecting licenses by full-text matching, with output in JSON, SPDX, CycloneDX and more, for open-source compliance reviews.

Runs open image models like FLUX, Qwen Image and Z-Image locally on a Mac: text-to-image, reference-based edits, inpainting and upscaling, all on-device. Needs Apple Silicon and uv; the first run downloads models and is slow. Check each model's license before commercial use.

Simon Willison's website screenshot CLI: batch screenshots from YAML, record demo videos, export PDF and HTML, run JavaScript in pages to extract data, and dump accessibility trees. Built on Playwright; install a browser on first use.

Translates whole EPUB books, Word files, SRT subtitles and TXT with local Ollama or cloud LLMs (OpenAI, Gemini, DeepSeek), keeping layout and timings. Builds a glossary for consistent terms and can resume jobs. Mainly a desktop app; the CLI runs from source; cloud models need an API key.

Apify's official skill collection: picks from 130+ curated Actors to scrape public data from social networks, maps, e-commerce sites and more, and can also build, package and deploy Actors. Needs an Apify account; a free tier is available.

Four tools for real-time web search, page content extraction, site mapping and full-site crawling, available as a remote server or run locally. Maintained by Tavily; needs an API key or OAuth sign-in, and a free account is available.

Translates entire books: splits PDF, DOCX or EPUB into chunks for parallel subagents, keeps terminology consistent, resumes interrupted runs and verifies with hashes, then outputs HTML, DOCX, EPUB or PDF. Needs Calibre, Pandoc and Python.